Image Processing Apparatus and Image Processing Method

This application is based upon and claims the benefit of priority from Japanese patent application No. 2025-006110, filed on Jan. 16, 2025, the disclosure of which is incorporated herein in its entirety by reference for all purposes.

The present disclosure relates to an image processing apparatus and an image processing method.

[Patent Literature 1] Japanese Patent No. 6249513 [Patent Literature 2] Japanese Unexamined Patent Application Publication No. 2008-032827 [Non-Patent Literature 1] Masaki Hayashi, “CVML expert guide”, [online], [searched on Jan. 7, 2025], Internet <URL: https://cvml-expertguide.net/2021/07/12/spatial-filtering/> Patent Literature 1 and 2 disclose techniques for inspecting a defect of a mask on which a pattern is formed.

Incidentally, it may be preferable to determine whether a defect candidate found on a pattern surface on which a pattern is formed should be detected as abnormal in consideration of an influence when the pattern surface is transferred to a wafer. Patent Literature 2 discloses that a pattern including a defect is inspected and corrected in an apparatus having optical characteristics the same as those of an exposure apparatus.

It is difficult, however, to separately create an inspection apparatus in accordance with optical conditions of the exposure apparatus.

The present disclosure has been made in view of the aforementioned problem, and provides an image processing apparatus and an image processing method capable of evaluating a defect candidate found on a pattern surface in consideration of an influence when the pattern surface is transferred to a wafer with a simple configuration.

An image processing apparatus according to one aspect of this embodiment includes: a first acquisition unit configured to acquire an evaluation image which is based on a result of imaging a pattern surface on which a pattern of an evaluation target mask is formed; a second acquisition unit configured to acquire a conversion parameter which is based on a relation between a captured image of the pattern surface and a transfer image when the pattern surface is transferred to a wafer in an exposure apparatus; and a third acquisition unit configured to acquire a converted evaluation image by applying the conversion parameter to the evaluation image.

In the above image processing apparatus, the second acquisition unit may acquire a lithography condition including an optical condition and a resist condition of the exposure apparatus and determine the conversion parameter based on the lithography condition.

In the above image processing apparatus, the second acquisition unit may train and determine the conversion parameter by machine learning which uses the captured image and the transfer image as a training data set.

In the above image processing apparatus, the second acquisition unit may determine, as the conversion parameter, a kernel which makes the captured image coincide with the transfer image when the kernel is applied to the captured image.

In the above image processing apparatus, the evaluation image may acquire the captured image, and the third acquisition unit may acquire, as the converted evaluation image, a converted captured image obtained by performing convolution integration in which the captured image is multiplied by the kernel.

In the above image processing apparatus, the second acquisition unit may determine the conversion parameter based on a function in an optical model that acquires information of the pattern surface from the captured image and a function in an optical model that acquires the transfer image from the information of the pattern surface.

In the above image processing apparatus, the first acquisition unit may acquire a difference image indicating a difference between the captured image and a reference image corresponding to the captured image as the evaluation image, and the third acquisition unit may acquire a converted difference image obtained by applying the conversion parameter to the difference image as the converted evaluation image.

The above image processing apparatus may further include a determination unit configured to determine whether the difference image satisfies a predetermined condition, in which when the determination unit determines that the predetermined condition is satisfied, the third acquisition unit may acquire the converted difference image as the converted evaluation image.

The above image processing apparatus may include an evaluation unit configured to determine that there is a defect in the evaluation target mask when a predetermined defective image is detected in the converted difference image.

The above image processing apparatus may further include a determination unit configured to determine classification of a defective image shown in the difference image, in which the second acquisition unit may acquire, in accordance with the classification determined by the determination unit, a correction conversion parameter obtained by correcting the conversion parameter as the conversion parameter, and the third acquisition unit may acquire, as the converted evaluation image, a difference image after correction conversion obtained by applying the correction conversion parameter to the difference image.

The above image processing apparatus may further include a determination unit configured to determine a defective image shown in the converted difference image as a defect in a specific classification.

In the above image processing apparatus, the second acquisition unit may acquire another parameter different from the conversion parameter as a conversion parameter, the third acquisition unit may acquire a difference image after other conversion obtained by applying the other parameter to the difference image, and the determination unit may determine a defective image shown in the converted difference image as a defect of first classification and determine a defective image shown in the difference image after other conversion as a defect of second classification.

The above image processing apparatus may further include a determination unit configured to determine whether the evaluation target mask satisfies a predetermined condition, in which when the determination unit determines that the predetermined condition is satisfied, the third acquisition unit may acquire the converted evaluation image.

In the above image processing apparatus, the predetermined condition may include at least one of a condition that the pattern surface of the evaluation target mask is illuminated in a specific illumination distribution in the exposure apparatus and a condition that the pattern surface has a specific shape.

An image processing method according to one aspect of this embodiment includes: a step of acquiring an evaluation image which is based on a result of imaging a pattern surface on which a pattern of an evaluation target mask is formed; a step of acquiring a conversion parameter which is based on a relation between a captured image of the pattern surface and a transfer image when the pattern surface is transferred to a wafer in an exposure apparatus; and a step of acquiring a converted evaluation image by applying the conversion parameter to the evaluation image.

In the above image processing method, in the step of acquiring the conversion parameter, a lithography condition including an optical condition and a resist condition of the exposure apparatus may be acquired and the conversion parameter may be determined based on the lithography condition.

In the above image processing method, in the step of acquiring the conversion parameter, the conversion parameter may be trained and determined by machine learning which uses the captured image and the transfer image as a training data set.

In the above image processing method, in the step of acquiring the conversion parameter, a kernel which makes the captured image coincide with the transfer image when the kernel is applied to the captured image may be determined as the conversion parameter.

In the above image processing method, the evaluation image may include the captured image, and in the step of acquiring the converted evaluation image, a converted captured image obtained by performing convolution integration in which the captured image is multiplied by the kernel may be acquired as the converted evaluation image.

In the above image processing method, in the step of acquiring the conversion parameter, the conversion parameter may be determined based on a function in an optical model that acquires information of the pattern surface from the captured image and a function in an optical model that acquires the transfer image from the information of the pattern surface.

In the above image processing method, in the step of acquiring the evaluation image, a difference image indicating a difference between the captured image and a reference image corresponding to the captured image may be acquired as the evaluation image, and in the step of acquiring the converted evaluation image, a converted difference image obtained by applying the conversion parameter to the difference image may be acquired as the converted evaluation image.

The above image processing method may further include a step of determining whether the difference image satisfies a predetermined condition, in which when it is determined, in the step of determining, that the predetermined condition is satisfied, in the step of acquiring the converted evaluation image, the converted difference image may be acquired as the converted evaluation image.

The above image processing method may include a step of determining that there is a defect in the evaluation target mask when a predetermined defective image is detected in the converted difference image.

The above image processing method may further include a step of determining classification of a defective image shown in the difference image, in which in the step of acquiring the conversion parameter, a correction conversion parameter obtained by correcting the conversion parameter may be acquired as the conversion parameter in accordance with the determined classification, and in the step of acquiring the converted evaluation image, a difference image after correction conversion obtained by applying the correction conversion parameter to the difference image may be acquired as the converted evaluation image.

The above image processing method may further include a step of determining a defective image shown in the converted difference image as a defect in a specific classification.

In the above image processing method, in the step of acquiring the conversion parameter, another parameter different from the conversion parameter may be acquired as the conversion parameter, in the step of acquiring the converted evaluation image, a difference image after other conversion obtained by applying the other parameter to the difference image may be acquired, and in the step of determining, a defective image shown in the converted difference image may be determined as a defect of first classification and a defective image shown in the difference image after other conversion may be determined as a defect of second classification.

The above image processing method may further include a step of determining whether the evaluation target mask satisfies a predetermined condition, in which when it is determined, in the step of determining, that the predetermined condition is satisfied, in the step of acquiring the converted evaluation image, the converted evaluation image may be acquired.

In the above image processing method, the predetermined condition may include at least one of a condition that the pattern surface of the evaluation target mask is illuminated in a specific illumination distribution in the exposure apparatus and a condition that the pattern surface has a specific shape.

According to the present disclosure, it is possible to provide an image processing apparatus and an image processing method capable of evaluating a defect candidate found on a pattern surface in consideration of an influence when the pattern surface is transferred to a wafer with a simple configuration.

The above and other objects, features and advantages of the present disclosure will become more fully understood from the detailed description given hereinbelow and the accompanying drawings.

Embodiments according to the present disclosure will be described hereinafter with reference to the drawings. In the following description, embodiments according to the present disclosure are shown, and the scope of the present disclosure is not limited to the below-shown embodiments. In the following description, components/structures to which the same reference numerals (or symbols) are assigned are substantially the same as each other.

An image processing apparatus and an image processing method according to a first embodiment will be described. The image processing apparatus according to this embodiment may be an evaluation apparatus that evaluates an evaluation target. The evaluation target includes, for example, a photomask. The photomask of the evaluation target is referred to as an evaluation target mask. The photomask may be the one on which a pattern is formed. The evaluation apparatus may include, as the evaluation target, not only a photomask but also a semiconductor substrate, a semiconductor chip, and a semiconductor device. The image processing apparatus may be a part of an optical apparatus such as an inspection apparatus. Further, the image processing apparatus may be a part of an apparatus (review apparatus) configured to display an obtained image (e.g., captured image) on a display or the like as a result of illumination of the evaluation target. Further, the image processing apparatus may be formed as a plurality of apparatuses such as a server, a client, and so on communicate with one another in a collaborative manner.

1 FIG. 1 FIG. 1 FIG. 10 10 11 12 13 11 12 13 10 14 14 is a block diagram illustrating an image processing apparatusaccording to the first embodiment. As shown in, the image processing apparatusincludes a first acquisition unit, a second acquisition unit, and a third acquisition unit. The first acquisition unit, the second acquisition unit, and third acquisition unitinclude functions as acquisition means. As shown in, the image processing apparatusmay include an evaluation unit. The evaluation unitincludes a function as evaluation means. The following description will be given using an evaluation target mask as an example of the evaluation target.

2 4 FIGS.- 3 FIG. 4 FIG. 5 FIG. 2 4 FIGS.- 1 11 10 1 1 1 1 1 11 1 10 11 1 52 51 50 are diagrams illustrating an evaluation image Aacquired by the first acquisition unitin the image processing apparatusaccording to the first embodiment.shows a captured image Bas the evaluation image Aandshows a difference image Cas the evaluation image A.is a diagram illustrating a reference image Dthat the first acquisition unituses to acquire the difference image Cin the image processing apparatusaccording to the first embodiment. As shown in, the first acquisition unitacquires an evaluation image Awhich is based on a result of imaging a pattern surfaceon which a patternof an evaluation target maskis formed.

3 FIG. 4 FIG. 5 FIG. 11 1 1 52 50 1 1 11 1 1 1 1 1 60 1 1 1 1 1 1 1 1 50 1 1 50 1 1 1 1 1 1 1 1 1 1 1 1 As shown in, the first acquisition unitmay acquire, as the evaluation image A, a captured image Bobtained by imaging the pattern surfaceof the evaluation target mask. That is, the evaluation image Amay include the captured image B. Further, as shown in, the first acquisition unitmay acquire the difference image Cas the evaluation image A. That is, the evaluation image Amay include the difference image C. The difference image Cmay include a defective image. The difference image Cis an image which is based on comparison between the captured image Band the reference image D(see) corresponding to the captured image B, and is, as an example, an image indicating the difference between the captured image Band the reference image Dcorresponding to the captured image B. The reference image Dmay be a non-defective image which shows an image in a case where there is no abnormality in an area of the evaluation target maskshown in the captured image Bor a case where there is no critical abnormality therein. Alternatively, the reference image Dmay be a comparison image which shows an area whose structure is equivalent to that of the area of the evaluation target maskshown in the captured image B. The difference image Cmay include, as pixel information, a difference in the luminance between pixels whose relative positions are the same among the pixels of the captured image Band the pixels of the reference image D. The difference image Cmay include, as pixel information, a ratio of the luminance of pixels whose relative positions are the same among the pixels of the captured image Band the pixels of the reference image D. Hereinafter, it is assumed that the difference image Cis an image indicating the difference between the captured image Band the reference image Dcorresponding to the captured image B, and pixel information of the difference image Cis simply called luminance.

6 FIG. 7 FIG. 1 12 10 1 12 10 1 is a diagram illustrating the captured image Bthat the second acquisition unituses to acquire conversion parameters in the image processing apparatusaccording to the first embodiment.is a diagram illustrating a wafer image Ethat the second acquisition unituses to acquire conversion parameters in the image processing apparatusaccording to the first embodiment. The wafer image Eis an example of a transfer image.

6 7 FIGS.and 12 1 52 50 1 1 52 51 50 1 51 50 54 1 55 51 50 54 As shown in, the second acquisition unitacquires conversion parameters which are based on a relation between the captured image Bof the pattern surfaceof the evaluation target maskand the wafer image E. Here, as described above, the captured image Bis an image obtained by imaging the pattern surfaceon which the patternof the evaluation target maskis formed. The wafer image Eis an image when the patternof the evaluation target maskis transferred to the waferin an exposure apparatus. The wafer image Eincludes an image of a patternobtained by transferring the patternof the evaluation target maskto the wafer.

1 1 1 1 The conversion parameters may include first to fourth parameters that will be described below. Further, the conversion parameters may include a fifth parameter. The fifth parameter includes a parameter other than the first to fourth parameters which is based on the relation between the captured image Band the wafer image E. For example, the fifth parameter may include other parameters that will be described later. Note that the conversion parameters may include correction conversion parameters obtained by correcting conversion parameters, as will be described later. The correction conversion parameters are also based on the relation between the captured image Band the wafer image E.

12 In the following, the conversion parameters acquired by the second acquisition unitwill be described separately in <First Parameter: Lithography Conditions>, <Second Parameter: Machine Learning Model>, <Third Parameter: Kernel>, and <Fourth Parameter: Functions of Fourier Transform and Inverse Fourier Transform>

12 12 12 The second acquisition unitmay acquire lithography conditions including optical conditions and resist conditions of the exposure apparatus. Then the second acquisition unitmay determine a conversion parameter (first parameter) based on the acquired lithography conditions. For example, in the exposure apparatus, the magnification in one direction (e.g., vertical direction) of the mask may be different from the magnification in another direction (e.g., lateral direction) perpendicular to one direction so as to correspond to high NA. In this case, as an example of the optical conditions, the magnification in the other direction with respect to one direction is M-fold. In this case, the second acquisition unitacquires, as a conversion parameter, a first parameter in which the magnification in the other direction with respect to one direction is M-fold.

51 Note that the optical conditions may include not only the magnification in the other direction with respect to one direction but also a luminance distribution, a contrast distribution or the like of illumination light. Further, the resist conditions may include a width between the patternsformed by resist, and so on.

12 1 1 12 12 1 1 12 The second acquisition unitmay train and determine a conversion parameter (second parameter) by machine learning which uses the captured image Band the wafer image Eas a training data set. For example, the second acquisition unitmay include a machine learning model. The second acquisition unitcauses the machine learning model to learn the captured image Band the wafer image Eas a training data set. Accordingly, the second acquisition unitmay acquire the trained machine learning model as the second parameter.

1 1 1 12 1 1 1 The second acquisition unit may determine, as a conversion parameter (third parameter), a coefficient of a kernel calculated in such a way that a captured image Bcoincides with a wafer image Eby performing convolution integration in which the captured image Bis multiplied by the kernel. The second acquisition unitmay determine, as a conversion parameter, a kernel that makes the captured image Bcoincide with the wafer image E(transfer image) when applied to the captured image B.

8 FIG. 8 FIG. 12 10 12 1 1 1 is a schematic view illustrating a kernel W that the second acquisition unitacquires as a conversion parameter in the image processing apparatusaccording to the first embodiment. As shown in, the second acquisition unitperforms convolution integration in which the captured image Bis multiplied by the kernel W. The captured image Bis an image formed of a plurality of pixel rows aligned in a first direction and a plurality of pixel rows aligned in a second direction perpendicular to the first direction. It is assumed, in this example, that the captured image Bis an image in which M pixels p are aligned in the first direction and N pixels p are aligned in the second direction.

3 1 It is assumed that the kernel W having two-dimensional parameters is, for example, a kernel havingrows and 3 columns, i.e., 3×3=9 cells. The size of each element (this may be referred to as a coefficient or a cell) of the kernel W may virtually correspond to the size of the pixel p of the captured image B.

12 1 12 1 3 1 The second acquisition unitperforms processing for applying the kernel W to the captured image B. For example, the second acquisition unitperforms convolution processing disclosed in Non-Patent Literature 1. Specifically, in the pixel p corresponding to the center of the kernel W, convolution processing is performed between surrounding 3×3 pixel values (e.g., luminance) and each cell of the kernel W to filter each pixel in the captured image B. For example, each cell of the kernel W is expressed by (i, j) using local coordinate axes including x and y axes. When a local operator for performing image filtering in neighboring× 3 pixels of each pixel p is denoted by J (p), J (p) can be defined as shown in the following Expression (1) as convolution between the luminance value I(x+i, y+j), which is around the pixel p=(x, y) on the captured image B, and W′ (p) (convolution is a two-dimensional discrete approximation value).

i, j 1 12 1 12 1 1 12 1 Here, W′ (p) is a kernel obtained by inverting the original kernel W (p) with respect to both x and y axes of the local coordinate axes. Further, w′is a weighting coefficient at each position (i, j) of the inverted kernel W′ (p), and S>0 is a coefficient for normalization and scaling. When i and j extend beyond the boundary range of the captured image B, padding processing may be performed first and a pixel value I(x+i, y+j) obtained by adding values to an area around the above position may be used. In this manner, the second acquisition unitperforms convolution processing on each pixel of the captured image Bto output pixel values after processing. The second acquisition unitcan sweep the whole captured image Bby the kernel W by sequentially changing pixels of the captured image Bwhere the central cell of the kernel W is arranged. Accordingly, the second acquisition unitacquires a corrected captured image to which the kernel W is applied for the whole captured image B.

1 1 m, n m, n m, n m, n Just like the captured image B, the corrected captured image is an image in which M pixels are aligned in the first direction and N pixels are aligned in the second direction. The luminance I'in each pixel (p′) of the corrected captured image is changed from the luminance Iof each pixel (p) of the captured image Bby the application of the kernel W.

12 1 12 1 The second acquisition unitacquires, as a third parameter, a kernel W in which the value of each cell (coefficient) is adjusted in such a way that the corrected captured image coincides with the wafer image E. For example, the second acquisition unitmay specify a coefficient in which a difference between the luminance of each pixel of the corrected captured image when the value of each cell (coefficient) of the kernel W is changed and applied and the luminance of each pixel of the wafer image Ebecomes minimum and determine the kernel W based on this coefficient.

12 52 1 1 52 The second acquisition unitmay determine a conversion parameter based on a function in an optical model MI for acquiring information of the pattern surfacefrom the captured image Band a function in an optical model ML for acquiring the wafer image Efrom information of the pattern surface.

9 9 FIGS.A andB 9 9 FIGS.A andB 9 FIG.A 9 FIG.B 12 10 1 52 50 50 are a diagram illustrating an optical model that the second acquisition unituses to acquire conversion parameters in the image processing apparatusaccording to the first embodiment. With reference to, the description will be made using a one-dimensional transmission optical model for the sake of explanation. The actual conversion parameters are acquired by expanding a one dimension to a two dimension. When an optical model that is different from the transmission optical model is used as well, descriptions similar to those shown below can be applied.shows an optical model MI of an optical apparatus (the optical apparatus may be an evaluation apparatus: hereinafter this apparatus will be referred to as an inspection apparatus) for acquiring the captured image Bof the pattern surfaceof the evaluation target mask.shows an optical model ML of the exposure apparatus for performing a lithography process using the evaluation target maskas a patterning device.

9 FIG.A 561 58 52 50 591 52 59 56 58 57 561 58 As shown in, the optical model MI of the inspection apparatus includes an optical elementand an optical elementI disposed between a pattern surfaceof an evaluation target maskand an image surface. Illumination light reflected by the pattern surfaceis concentrated on the image surfaceI via the optical elementI and the optical elementI. A pupil surfaceI is formed between the optical elementand the optical elementI.

9 FIG.A 591 571 52 In, a one-dimensional position is denoted by x and an angular distribution is denoted by k. The strength of the image surface, the strength of the pupil surface, and the strength of the pattern surfaceare respectively expressed by the following Expressions (2), (3), and (4). Here, i is an imaginary number.

52 571 561 57 591 581 Here, uI(k) is a function that depends on the optical design of the inspection apparatus for reproducing the change in the strength from the pattern surfaceto the pupil surfaceby the optical elementof the inspection apparatus. Further, vI(k) is a function that depends on the optical design of the inspection apparatus for reproducing the change in the strength from the pupil surfaceI to the image surfaceby the optical elementof the inspection apparatus.

1 1 52 50 1 gI(x) correspond to one-dimensional(x-direction) strength of the captured image B. Therefore, by performing, based on the optical model MI of the inspection apparatus, Fourier transform shown in Expression (3) on gI(x) indicating the strength of the captured image Band performing inverse Fourier transform shown in Expression (4) on the obtained FI(k), the strength of the pattern surfaceof the evaluation target mask(this may correspond to a physically uneven shape) can be calculated from the captured image B.

9 FIG.B 56 58 52 50 59 52 59 56 58 57 56 58 As shown in, the optical model ML of the exposure apparatus includes an optical elementL and an optical elementL disposed between a pattern surfaceof an evaluation target mask(as an application, a mask as a patterning device) and an image surfaceL. Illumination light reflected by the pattern surfaceis concentrated on the image surfaceL via the optical elementL and the optical elementL. A pupil surfaceL is formed between the optical elementL and the optical elementL.

9 FIG.B 52 57 59 59 1 In, a one-dimensional position is denoted by x and an angular distribution is denoted by k. When the strength of the pattern surfaceis denoted by the above f(x), the strength of the pupil surfaceL and the strength of the image surfaceL are respectively expressed by the following Expressions (5) and (6). Here, the strength gL(x) of the image surfaceL corresponds to one-dimensional (x-direction) strength of the wafer image E.

52 57 56 57 59 58 Here, uL(k) is a function that depends on the optical design of the exposure apparatus for reproducing the change in the strength from the pattern surfaceto the pupil surfaceL by the optical elementL of the exposure apparatus. Further, vL(k) is a function that depends on the optical design of the exposure apparatus for reproducing the change in the strength from the pupil surfaceL to the image surfaceL by the optical elementL of the exposure apparatus.

1 52 50 f(x) is acquired based on the aforementioned Expression (4). Therefore, the strength of the wafer image Ecan be calculated from the strength f(x) of the pattern surfaceof the evaluation target maskby performing Fourier transform shown in Expression (5) for f(x) based on the optical model ML of the exposure apparatus and performing inverse Fourier transform shown in Expression (6) on the obtained FL(k).

52 50 1 1 1 To sum up, by further performing the Fourier transform shown in Expression (5) and the inverse Fourier transform shown in Expression (6) which are based on the optical model ML of the exposure apparatus on the strength of the pattern surfaceof the evaluation target maskobtained by performing the Fourier transform shown in Expression (3) and the inverse Fourier transform shown in Expression (4) which are based on the optical model MI of the inspection apparatus on gI(x) indicating the strength of the captured image B, the wafer image Ecan be calculated and acquired from the captured image B.

12 1 1 12 52 1 1 52 The second acquisition unitacquires a fourth parameter for converting the captured image Binto a wafer image Ebased on the optical model. Specifically, the second acquisition unitdetermines the fourth parameter based on a function in the optical model MI for acquiring information of the pattern surfacefrom the captured image Band a function in the optical model ML for acquiring the wafer image Efrom information of the pattern surface. The fourth parameter may be a parameter that is determined based on the above uI(k), vI(k), uL(k), and vL(k), or may be a parameter that includes the above uI(k), vI(k), uL(k), and vL(k) as an internal coefficient.

10 FIG. 10 FIG. 1 13 10 13 1 1 1 1 1 11 1 1 1 1 1 1 is a diagram illustrating a converted evaluation image Facquired by the third acquisition unitin the image processing apparatusaccording to the first embodiment. As shown in, the third acquisition unitacquires the converted evaluation image Fby applying conversion parameters to an evaluation image A. The evaluation image Aincludes a captured image Band a difference image C. The first acquisition unitmay acquire, as the difference image C, an image indicating a difference between the captured image Band the converted reference image obtained by applying conversion parameters to the reference image D. Therefore, the evaluation image Amay include the difference image Cbetween the captured image Band the converted reference image.

The conversion parameters include the first to fifth parameters and the correction conversion parameters.

1 1 1 13 1 1 1 1 1 1 13 1 1 1 1 1 1 1 1 1 The converted evaluation image Fincludes a converted captured image Gand a converted difference image H. For example, the third acquisition unitmay acquire, as the converted evaluation image F, a converted captured image Gobtained by applying conversion parameters to the captured image B(e.g., performing convolution integration in which the captured image Bis multiplied by the kernel W). At this time, the luminance of at least some of the pixels included in the converted captured image Gis changed from the luminance of the pixels included in the captured image Bbased on the application of the conversion parameters. Further, the third acquisition unitmay acquire, as the converted evaluation image F, a converted difference image Hobtained by applying conversion parameters to the difference image C(e.g., performing convolution integration in which the difference image Cis multiplied by the kernel W). At this time, the luminance of at least some of the pixels included in the converted difference image His changed from the luminance of pixels included in the difference image Cbased on the application of the conversion parameters. The converted evaluation image Fmay include a difference image after correction conversion Jand a difference image after other conversion Kthat will be described later.

14 52 50 1 14 52 50 1 1 1 1 14 50 60 1 1 1 1 1 1 1 1 The evaluation unitevaluates the pattern surfaceof the evaluation target maskbased on the converted evaluation image F. Specifically, the evaluation unitevaluates the pattern surfaceof the evaluation target maskbased on the converted captured image G, the converted difference image H, the difference image after correction conversion J, and the difference image after other conversion K. For example, the evaluation unitmay determine that there is a defect in the evaluation target maskwhen a predetermined defective imagehas been detected in the converted difference image Hobtained by applying a conversion parameter to the difference image Cbetween the captured image Band the reference image D. Here, the evaluation image Ais the difference image C, and the converted evaluation image Fis the converted difference image H.

14 50 60 1 1 1 1 1 1 1 1 14 50 1 1 1 1 14 52 50 1 Further, the evaluation unitmay determine that there is a defect in the evaluation target maskwhen a predetermined defective imagehas been detected in the difference image Cbetween the converted captured image Gobtained by applying conversion parameters to the captured image Band the reference image D(it may be a converted reference image). Here, the evaluation image Ais the captured image B, and the converted evaluation image Fis the converted captured image G. In this manner, a configuration in which the evaluation unitdetermines a defect of the evaluation target maskbased on the difference image Cindicating the difference between the converted captured image G, which is the converted evaluation image F, and the reference image D(it may be a converted reference image) is also included in a configuration in which the evaluation unitevaluates the pattern surfaceof the evaluation target maskbased on the converted evaluation image F.

60 The aforementioned predetermined defective imagemay include a defective image including pixels with luminance outside a predetermined range, a defective image including a predetermined number or more of pixels with luminance outside the predetermined range, and so on.

11 FIG. 11 FIG. 11 FIG. 11 1 12 13 1 14 Next, an image processing method according to this embodiment will be described.is a flowchart illustrating the image processing method according to the first embodiment. As shown in, the image processing method according to this embodiment includes Step Sof acquiring an evaluation image A, Step Sof acquiring conversion parameters, and Step Sof acquiring a converted evaluation image F. As shown in, the image processing method according to this embodiment may further include Step Sof performing evaluation.

11 11 1 52 51 50 1 1 1 11 1 1 1 1 1 11 1 1 1 1 In Step S, the first acquisition unitacquires the evaluation image Awhich is based on a result of imaging the pattern surfaceon which the patternof the evaluation target maskis formed. The evaluation image Amay include a captured image Bor may include a difference image C. The first acquisition unitmay acquire, as the evaluation image A, the difference image Cindicating a difference between the captured image Band the reference image Dcorresponding to the captured image B. Further, the first acquisition unitmay acquire, as the difference image C, an image indicating a difference between the captured image Band a converted reference image obtained by applying conversion parameters to the reference image Dcorresponding to the captured image B.

12 12 1 1 12 12 1 1 12 1 1 1 12 52 1 1 52 In Step S, the second acquisition unitacquires conversion parameters which are based on the relation between the captured image Band the wafer image E. The second acquisition unitmay acquire lithography conditions including optical conditions and resist conditions of the exposure apparatus, and determine conversion parameters based on the lithography conditions. Further, the second acquisition unitmay train and determine conversion parameters by machine learning that uses the captured image Band the wafer image Eas a training data set. Further, the second acquisition unitmay determine, as a conversion parameter, a coefficient of a kernel calculated in such a way that the captured image Bcoincides with the wafer image Eby performing convolution integration in which the captured image Bis multiplied by the kernel. Furthermore, the second acquisition unitmay determine conversion parameters based on a function in the optical model MI for acquiring information of the pattern surfacefrom the captured image Band a function in the optical model ML for acquiring the wafer image Efrom information of the pattern surface.

13 13 1 1 13 1 1 1 13 1 1 1 In Step S, the third acquisition unitapplies conversion parameters to the evaluation image Ato acquire a converted evaluation image F. For example, the third acquisition unitmay acquire, as the converted evaluation image F, a converted captured image Gobtained by performing convolution integration in which a kernel is multiplied by the captured image B. Further, the third acquisition unitmay acquire, as the converted evaluation image F, a converted difference image Hobtained by applying conversion parameters to the difference image C.

14 14 52 1 In Step S, the evaluation unitevaluates the pattern surfaceof the evaluation target mask based on the converted evaluation image F.

10 1 52 50 1 52 54 10 52 50 1 1 10 52 52 54 Next, effects of this embodiment will be described. The image processing apparatusaccording to this embodiment acquires conversion parameters which are based on a relation between the captured image Bof the pattern surfaceof the evaluation target maskand the wafer image Ewhen the pattern surfaceis transferred to the waferin an exposure apparatus. Then, the image processing apparatusevaluates the pattern surfaceof the evaluation target maskbased on a converted evaluation image Fobtained by applying conversion parameters to the evaluation image A. Accordingly, the image processing apparatuscan evaluate a defect candidate found on the pattern surfacein consideration of an influence when the pattern surfaceis transferred to the waferwith a simple configuration.

1 1 54 54 10 1 1 54 1 50 54 When a detect or a defect candidate shown on the captured image Bor the difference image Cis transferred to the waferin the exposure apparatus, it is possible that they do not give an adverse effect on a product formed on the wafer. The image processing apparatusevaluates whether the defect and the defect candidate shown in the captured image Bor the difference image Cmay give an adverse effect on the product formed on the waferby the converted evaluation image Fconverted by conversion parameters in advance. Accordingly, it is possible to evaluate whether a defect or a defect candidate gives an adverse effect even when the evaluation target maskis not actually transferred to the waferin the exposure apparatus.

12 10 12 50 The second acquisition unitof the image processing apparatusmay determine conversion parameters based on lithography conditions of the exposure apparatus, or may train and determine conversion parameters by machine learning. Further, the second acquisition unitmay determine a coefficient of a kernel W as a conversion parameter, or may determine a conversion parameter based on functions of Fourier transform and inverse Fourier transform of the optical model. It is therefore possible to improve the accuracy of evaluating the evaluation target mask.

60 20 20 10 20 11 11 14 11 11 14 12 FIG. 12 FIG. a a a a Next, an image processing apparatus according to a second embodiment will be described. The image processing apparatus according to this embodiment includes a determination unit configured to determine predetermined conditions and classification of defective images.is a block diagram illustrating an image processing apparatusaccording to the second embodiment. As shown in, the image processing apparatusis different from the above-described image processing apparatusin that the image processing apparatusfurther includes a determination unit. The determination unitincludes a function as determination means. An evaluation unitmay include the determination unit. The determination unitmay include the evaluation unit.

11 1 52 1 52 52 52 51 11 1 12 13 1 11 50 11 50 12 13 1 a a a a The determination unitdetermines whether an evaluation image Asatisfies predetermined conditions. The predetermined conditions include, for example, at least one of a condition that a pattern surfacein a captured image Bhas a specific shape and a condition that the pattern surfaceis illuminated in a specific illumination distribution in an exposure apparatus. The pattern surfacehaving a specific shape includes, for example, a case in which the pattern surfacehas a line shape such as line-and-space, and a hole shape. Illumination in a specific illumination distribution includes, for example, an illumination distribution in a dual-polarized light source. When a patternof the line and space is exposed, illumination from a dual-polarized light source such as a dipole may be used in order to increase the contrast. In this embodiment, when the determination unitdetermines that the evaluation image Asatisfies the predetermined conditions, a second acquisition unitacquires conversion parameters and a third acquisition unitacquires a converted evaluation image F. Note that the determination unitmay determine whether an evaluation target masksatisfies predetermined conditions. When the determination unitdetermines that the evaluation target masksatisfies the predetermined conditions, the second acquisition unitmay acquire conversion parameters and the third acquisition unitmay acquire the converted evaluation image F.

11 1 60 1 11 1 12 13 1 1 14 52 50 1 a a Further, the determination unitmay determine whether a difference image Csatisfies predetermined conditions. The predetermined conditions may include a condition related to at least one of the size, the luminance, a change in the luminance, and the shape of the defective imagein the difference image C. When the determination unithas determined that the difference image Csatisfies predetermined conditions, the second acquisition unitacquires conversion parameters, and the third acquisition unitacquires a converted difference image Has the converted evaluation image F. The evaluation unitevaluates the pattern surfaceof the evaluation target maskbased on the converted difference image H.

20 12 13 14 1 1 1 20 12 13 14 Accordingly, the image processing apparatusoperates the second acquisition unit, the third acquisition unit, and the evaluation unitwhen the evaluation image Asuch as the captured image Band the difference image Csatisfies predetermined conditions. When the predetermined conditions are not satisfied, the image processing apparatusdoes not need to operate the second acquisition unit, the third acquisition unit, and the evaluation unit, whereby it is possible to reduce the cost.

60 1 1 1 1 60 50 Depending on the size, the luminance, a change in the luminance, and the shape of the defective imagein the difference image C, it may be preferable to apply conversion parameters using the evaluation image Fas the difference image Cand perform evaluation based on the converted difference image H. Accordingly, by setting the conditions such as the size, the luminance, a change in the luminance, and the shape of the defective image, it is possible to improve the accuracy of evaluating the evaluation target mask.

11 60 1 11 60 11 60 60 60 11 60 11 60 60 a a a a a Further, the determination unitmay determine classification of a defective imageshown in the difference image C. For example, the determination unitmay classify a change in the luminance in the area of the defective image. Specifically, the determination unitmay classify defective imagesinto defective imageswhose changes in the luminance are gentle and defective imageswhose changes in the luminance are steep. Further, the determination unitmay classify the size, the luminance, and the shape of the defective images. The determination unitmay classify the defective imageswhose changes in the luminance are gentle and the defective imageswhose changes in the luminance are steep depending on a threshold in a rate of change in the luminance.

13 14 FIGS.and 13 14 FIGS.and 13 FIG. 12 1 13 20 12 1 60 11 61 60 12 1 1 13 1 1 a are diagrams illustrating conversion parameters acquired by the second acquisition unitand the converted evaluation image Facquired by the third acquisition unitin the image processing apparatusaccording to the second embodiment. As shown in, the second acquisition unitacquires conversion parameters applied to the difference image Cin accordance with classification of the defective imagedetermined by the determination unit. For example, as shown in, in a case of a classificationwhere a change in luminance in an area of a defective imageis steep, the second acquisition unitacquires predetermined conversion parameters which are based on a relation between the captured image Band the wafer image E. The third acquisition unitthen applies conversion parameters to the difference image Cto acquire the converted difference image H.

14 FIG. 62 60 12 13 1 13 1 1 1 1 On the other hand, as shown in, in a case of a classificationin which a change in luminance in an area of a defective imageis gentle, the second acquisition unitacquires correction conversion parameters obtained by correcting conversion parameters as conversion parameters. Then, the third acquisition unitapplies the correction conversion parameters to the difference image Cas conversion parameters. Accordingly, the third acquisition unitacquires a difference image after correction conversion Jas the converted evaluation image F. The difference image after correction conversion Jis an image obtained by applying the correction conversion parameters to the difference image C.

12 60 12 60 60 62 1 1 62 60 62 12 60 52 50 60 For example, the second acquisition unitmay acquire the correction conversion parameters by assigning components of a moving averaging filter and/or a sharpening filter in accordance with the classification of the defective imageto conversion parameters. The second acquisition unitmay acquire the correction conversion parameters by assigning components of a filter for other purpose in accordance with the classification of the defective imageto the conversion parameters. It is possible that a defect that is detected as a defective imageof the classificationin which the change in the luminance is gentle in the difference image Cmay not be easily detected in the wafer image Eobtained as a result of transfer in an exposure apparatus. Then, it is possible that the influence of the defective image of the classificationmay not be appropriately determined. In order to solve this problem, in the case of the defective imageof the classificationin which the change in the luminance is gentle, correction conversion parameters obtained by further assigning components of the sharpening filter to the conversion parameters are used. As stated above, the second acquisition unitmay acquire conversion parameters in accordance with the classification of the defective images. Accordingly, it is possible to evaluate the pattern surfaceof the evaluation target maskin a state in which the influence of an area near the defective imageis eliminated.

11 60 60 60 60 50 54 60 61 60 62 60 60 1 13 1 1 60 1 13 1 1 a Note that the determination unitmay classify defective imagesbased on the size of the defective image, the shape of the defective image, the luminance of the defective image, the material of the evaluation target mask, the material of the wafer, or the like, not only based on the change in the luminance in the area of the defective image. Further, while conversion parameters are applied in the classificationin which the change in the luminance in the area of the defective imageis steep and correction conversion parameters are applied in the classificationin which the change in the luminance in the area of the defective imageis gentle in the above description, this relation may be reversed. In addition, the distinction of the classification is only one example. Therefore, when the defective imagein the difference image Cis a defect (defective image) of first classification, the third acquisition unitmay acquire the converted difference image Hobtained by applying conversion parameters to the difference image C. On the other hand, when the defective imagein the difference image Cis a defect (defective image) of the second classification, the third acquisition unitmay acquire a difference image after correction conversion Jobtained by applying correction conversion parameters to the difference image C.

15 FIG. 15 FIG. 12 1 13 20 13 1 1 13 1 1 is a diagram illustrating the conversion parameters acquired by the second acquisition unitand the converted evaluation image Facquired by the third acquisition unitin the image processing apparatusaccording to the second embodiment. As shown in, the third acquisition unitmay acquire a converted difference image Hby applying conversion parameters to the difference image C. Further, the third acquisition unitmay acquire a difference image after correction conversion Jby applying correction conversion parameters to the difference image C.

20 1 60 11 14 60 1 60 61 14 60 1 60 62 14 11 11 60 1 11 60 1 61 60 1 62 a a a a As described above, the image processing apparatusmay acquire two evaluation images Fwithout determining a defective imageby the determination unit. The evaluation unitmay determine a defective imagedetected in the converted difference image Hto which conversion parameters are applied as a defective imageof the classification. Further, the evaluation unitmay determine a defective imagedetected in the difference image after correction conversion Jto which correction conversion parameters are applied as a defective imageof the classification. As described above, the evaluation unitmay include the determination unit. Therefore, the determination unitmay determine that the defective imageshown in the converted difference image Has a defect in a specific classification. Further, the determination unitmay determine that the defective imageshown in the converted difference image Has a defect of the classificationand determine that the defective imageshown in the difference image after correction conversion Jas a defect of the classification.

61 62 60 60 60 60 20 60 11 15 FIG. a. By using conversion parameters in accordance with each of the classificationand classificationof the defective images, these defective imagesmay become more noticeable. That is, by applying conversion parameters, it is possible to characteristically extract defective imagesin a specific specification. Accordingly, by changing conversion parameters, a specific defective imagecan be selectively detected. Further, an image processing method performed by the image processing apparatusincan eliminate the need for preliminary classification of the defective imagesby the determination unit

12 13 1 1 1 13 1 1 1 1 1 1 11 60 1 1 a The second acquisition unitmay acquire other parameters different from the conversion parameters as conversion parameters. The third acquisition unitmay acquire, as a converted evaluation image F, a difference image after other conversion Kobtained by applying other parameters to the difference image C. Further, the third acquisition unitmay acquire a converted difference image Hby applying conversion parameters to the difference image Cand then acquire a correction conversion difference image Jby applying correction conversion parameters to the converted difference image H. In this case, the other parameters may be the ones obtained by operating conversion parameters and correction conversion parameters. The other parameters may be the same as the correction conversion parameters, and the correction conversion difference image Jmay be the same as the difference image after other conversion K. The determination unitmay determine that the defective imageshown in the converted difference image His a defect of the first classification and determine that the defective image shown in the difference image after other conversion Kis a defect of second classification.

16 FIG. 16 FIG. 11 11 11 1 11 1 11 60 1 a a a a a is a flowchart illustrating the image processing method according to the second embodiment. As shown in, the image processing method according to this embodiment is different from the aforementioned image processing method in that the image processing method according to this embodiment includes Step S. In Step S, the determination unitdetermines whether an evaluation image Asatisfies predetermined conditions. The determination unitmay determine whether a difference image Csatisfies predetermined conditions. Further, the determination unitmay determine classification of the defective imageshown in the difference image C.

11 1 1 12 12 12 60 1 11 a a. When it is determined by the determination unitthat the evaluation image Aor the difference image Csatisfies predetermined conditions, the second acquisition unitmay acquire conversion parameters in Step S. Further, the second acquisition unitacquires, for the defective imageshown in the difference image C, conversion parameters or correction conversion parameters in accordance with the classification determined by the determination unit

20 11 60 1 20 60 1 20 11 1 1 1 11 60 60 1 1 11 13 a a a a According to this embodiment, the image processing apparatusincludes the determination unitthat determines classification of the defective imagein the difference image C. Therefore, the image processing apparatuscan apply conversion parameters in accordance with the classification of the defective imageto the difference image C. It is therefore possible to improve the accuracy of evaluation. Further, since the image processing apparatusincludes the determination unitthat determines predetermined conditions, the evaluation image Acan be determined before conversion parameters are applied to the evaluation image A. It is therefore possible to determine whether or not to acquire the converted evaluation image F. It is therefore possible to reduce the processing cost. As described above, the determination unitmay determine the classification of the defective imagebased on the defective imagespecified based on the converted evaluation image Fobtained by applying specific conversion parameters to the evaluation image A. In this case, Step Smay be provided at a stage later than Step S.

50 Next, an optical apparatus according to a third embodiment will be described. The optical apparatus according to this embodiment includes an image processing apparatus. The optical apparatus may include, for example, an inspection apparatus for inspecting an evaluation target mask, and an image processing apparatus. Further, the optical apparatus may include a review apparatus and an image processing apparatus. Note that the image processing apparatus may be a part of the inspection apparatus and a part of the review apparatus, and the optical apparatus may be each of the inspection apparatus and the review apparatus themselves. In the following, as one example of the optical apparatus, an inspection apparatus including an image processing apparatus will be described.

17 FIG. 100 100 110 120 10 10 20 110 50 111 110 111 112 113 114 is a configuration diagram illustrating an inspection apparatusas an optical apparatus according to the third embodiment. The inspection apparatusincludes an illumination optical system, an imaging optical system, and an image processing apparatus. Note that the image processing apparatusmay be an image processing apparatus. The illumination optical systemilluminates an evaluation target maskusing illumination light L. The illumination optical systemincludes, for example, a light source, an elliptic mirror, an elliptic mirror, and a drop-in mirror.

120 1 50 111 120 121 122 123 121 122 110 120 The imaging optical systemcaptures a captured image Bof the evaluation target maskilluminated by the illumination light L. The imaging optical systemincludes, for example, a concave mirrorwith a hole formed therein, a convex mirror, and a detector. The concave mirrorwith the hole formed therein and the convex mirrorform a Schwarzschild magnification optical system. Note that the illumination optical systemand the imaging optical systemmay further include optical members other than those described above, and do not necessarily have to include all of the above-described optical members.

100 152 50 100 100 Here, for the sake of explanation of the inspection apparatus, an XYZ-orthogonal coordinate system is introduced. For example, a plane parallel to a stage surface of a stageon which the evaluation target maskis disposed is defined as an XY-plane, and a direction perpendicular to the stage surface is defined as a Z-axis direction. The +Z-axis direction is referred to as upward and the −Z-axis direction is referred to as downward for the sake of convenience. Note that the terms upward and downward are used only for the sake of explanation of the inspection apparatus, and they do not indicate directions in which the inspection apparatusis actually disposed.

111 111 111 50 111 111 112 111 112 1 1 52 50 The light sourcegenerates illumination light L. The illumination light Lcontains, for example, EUV light of 13.5 nm, which is equal to an exposure wavelength of a EUV mask, i.e., the evaluation target mask. The illumination light Lgenerated by the light sourceis reflected by the elliptic mirror. The illumination light Lreflected by the elliptic mirrortravels while being narrowed, and is concentrated at a concentration point IF. The concentration point IFis positioned at a position conjugate with a pattern surfaceof the evaluation target mask.

1 111 113 111 113 113 114 113 111 114 114 50 111 114 50 114 111 50 After passing through the concentration point IF, the illumination light Ltravels while being expanded, and is incident on a reflecting mirror such as the elliptic mirror. The illumination light Lincident on the elliptic mirroris reflected by the elliptic mirror, travels while being narrowed, and is incident on the drop-in mirror. That is, the elliptic mirrormakes the illumination light Lincident on the drop-in mirroras converging light. The drop-in mirroris disposed above the evaluation target mask. The illumination light Lincident on and reflected by the drop-in mirroris incident on the evaluation target mask. That is, the drop-in mirrormakes the illumination light Lincident on the evaluation target mask.

113 111 50 110 111 50 111 52 50 110 110 50 111 111 The elliptic mirrorconcentrates the illumination light Lon the evaluation target mask. The illumination optical systemis disposed so that when the illumination light Lilluminates the evaluation target mask, an image of the light sourceis formed on the pattern surfaceof the evaluation target mask. Therefore, the illumination optical systemprovides critical illumination. As described above, the illumination optical systemilluminates the evaluation target maskby using the critical illumination by the illumination light Lgenerated by the light source.

50 152 111 50 111 50 111 50 The evaluation target maskis disposed on the stage. The illumination light Lis incident on the evaluation target maskin a direction inclined from the Z-axis direction. That is, the illumination light Lis obliquely incident on the evaluation target maskas oblique incident illumination. The illumination light Lmay illuminate the evaluation target maskas the oblique incident illumination.

152 50 152 152 152 152 110 120 The stageis an XYZ-driven stage. It is possible to illuminate a desired area on the evaluation target maskby moving the stagein the X-axis and Y-axis directions. Further, it is possible to adjust the focus by moving the stagein the Z-axis direction. Further, the stagemay be rotated around the X-axis, Y-axis and Z-axis. Note that instead of moving and rotating the stagein the X-axis, the Y-axis, and Z-axis directions, the illumination optical systemand the imaging optical systemmay be moved and rotated.

111 111 50 112 50 50 121 121 121 a The illumination light Lemitted from the light sourceilluminates a captured area on the evaluation target mask. Reflected light L, i.e., light that has been incident on the evaluation target maskin a direction inclined from the Z-axis direction and reflected by the evaluation target mask, is incident on the concave mirrorwith the hole formed therein. A holeis formed at the center of the concave mirror.

112 121 122 122 112 121 121 121 112 121 123 123 123 a a The reflected light Lreflected by the concave mirrorwith the hole formed therein is incident on the convex mirror. The convex mirrorreflects the reflected light Lincident from the concave mirrorwith the hole formed therein toward the holeof the concave mirror. The reflected light Lpassing through the holeis detected by the detector. The detectormay be a detectorincluding a Time Delay Integration (TDI) sensor.

123 50 123 123 The detectoracquires image data of the evaluation target mask. The detectorincludes a plurality of photographing elements arranged linearly in one direction. Linear image data taken by the plurality of linearly-arranged photographing elements is referred to as one-dimensional image data or one frame. The detectoracquires a plurality of such one-dimensional image data by performing scanning in a direction perpendicular to the one direction. The photographing elements are, for example, a Charge Coupled Device(s) (CCDs). Note that the photographing elements are not limited to the CCD(s).

120 112 50 111 50 123 112 In this way, the imaging optical systemconcentrates the reflected light Lcoming from the evaluation target maskilluminated by the illumination light L, and acquires image data of the evaluation target maskby having the detectordetect the concentrated reflected light L. The image data is, for example, two-dimensional image data.

112 60 50 111 50 120 50 60 50 123 10 The reflected light Lcontains information about a defective imageor the like on the evaluation target mask. The normal reflected light of the illumination light Lthat has been incident on the evaluation target maskin the direction inclined from the Z-axis direction is detected by the imaging optical system. When there is a defect on the evaluation target mask, the defective imageis observed as a dark image. Such an observation method is referred to as bright-field observation. The plurality of one-dimensional image data of the evaluation target maskacquired by the detectorare output to the image processing apparatusand processed into two-dimensional image data.

10 120 10 50 123 120 10 50 123 1 10 1 1 The image processing apparatusis connected to the imaging optical systemby a signal line or wirelessly. The image processing apparatusreceives the image data of the evaluation target maskfrom the detectorin the imaging optical system. The image processing apparatusperforms image processing on the image data of the evaluation target maskreceived from the detectoras a two-dimensional captured image B. Specifically, the image processing apparatusconverts the captured image Binto a converted captured image Gbased on conversion parameters.

50 50 50 111 50 The evaluation target maskis, for example, an EUV mask sensitive to EUV light. Note that the evaluation target maskis not limited to the EUV mask. The evaluation target maskmay be a photomask sensitive to the illumination light Lhaving other wavelengths. Further, as described above, the evaluation target maskmay be a semiconductor substrate, a semiconductor chip, a semiconductor device, or the like.

While embodiments according to the present disclosure have been described above, the present disclosure includes modifications as appropriate without impairing the objects and advantages thereof, and is not limited by the above-described embodiments. Further, combinations of the configurations of the first to third embodiments and their modified examples are also within the scope of the technical concept of the present disclosure.

A program can be stored and provided to a computer using any type of non-transitory computer readable media. Non-transitory computer readable media include any type of tangible storage media. Examples of non-transitory computer readable media include magnetic storage media (such as flexible disks, magnetic tapes, hard disk drives, etc.), optical magnetic storage media (e.g. magneto-optical disks), CD-ROM (compact disc read only memory), CD-R, CD-R/W, and semiconductor memories (such as mask ROM, PROM (programmable ROM), EPROM (erasable PROM), flash ROM, RAM (random access memory), etc.). The program may be provided to a computer using any type of transitory computer readable media. Examples of transitory computer readable media include electric signals, optical signals, and electromagnetic waves. Transitory computer readable media can provide the program to a computer via a wired communication line (e.g. electric wires, and optical fibers) or a wireless communication line.

The first to third embodiments can be combined as desirable by one of ordinary skill in the art.

From the disclosure thus described, it will be obvious that the embodiments of the disclosure may be varied in many ways. Such variations are not to be regarded as a departure from the spirit and scope of the disclosure, and all such modifications as would be obvious to one skilled in the art are intended for inclusion within the scope of the following claims.